Advanced deinterlacer for high-definition and standard-definition video

Abstract

A motion adaptive video deinterlacer may process fields of video derived from frames of video. The deinterlacer may use multiple pixel motion engines to provide motion information about the pixels within each field. The output of the motion engines may be used to deinterlace the fields of video based on the detail within a field of video. The deinterlacer may use motion recursion and motion recirculation to provide temporal motion expansion for the pixels within each field. In addition, the deinterlacer may detect various cadences for various regions within the frames of video. The cadences may be detected using a calculated threshold, or without using a calculated threshold.

Description

CROSS REFERENCE TO RELATED APPLICATIONS [0001]This claims the benefit of copending, commonly-assigned U.S. Provisional Patent Application No. 60 / 864,881 filed Nov. 8, 2006, which is hereby incorporated by reference herein in its entirety.BACKGROUND OF THE INVENTION [0002]This invention generally relates to systems and methods for deinterlacing high-definition and standard definition video using adaptive techniques for spatial and temporal motion.[0003]Video is often interlaced in order to improve picture quality and decrease the bandwidth required for transmission. Interlaced video formats provide one half of the lines in an image for any given frame—the odd-numbered horizontal lines of the frame may be displayed at one instant, while the even-numbered lines of the frame may be displayed the next. Each set of lines can be referred to as a field. In order to display the video in progressive, or noninterlaced, format, interlaced video must be deinterlaced. Deinterlacing requires the d...

AI Technical Summary

Benefits of technology

The patent describes a video deinterlacer that can adapt to motion in the video frames. It uses pixel motion engines to create multiple delayed versions of the frames and their corresponding fields. The deinterlacer also includes motion handling circuitry to produce motion information about each pixel in the fields. The motion information is used to create temporal motion expansion of the fields by applying a weighted average function. The deinterlacer also includes cadence detection circuitry to detect cadences in the video and enter and exit a particular cadence mode of processing based on the calculated noise threshold or tracking the position of a minimum difference between groups of pixels. The technical effects of the patent are improved video quality and smoother video processing.

Problems solved by technology

Video-mode deinterlacing often results in undesirable artifacts in the video.

However, displaying video on televisions that are 42 inches or larger in size and capable of displaying resolutions of 1920 by 1080 pixels can cause these artifacts to be more visible.

This use of deinterlacing may cause the text to appear blurred, or feathered, and may introduce artifacts such as line flicker and Moire.

This deficiency may cause portions of video to be deinterlaced using suboptimal techniques.

Images